classes=classes,
weights=None,
dropout=dropout,
fine=1,
retrain=False,
pre_file=pre_file,
old_epochs=old_epochs,
cross_index=cross_index,
input_shape=spatial_a.output_shape,
input_tensor=spatial_a.output)
spatial = models.InceptionSpatial(n_neurons=n_neurons,
seq_len=seq_len,
classes=classes,
weights=None,
dropout=dropout,
fine=False,
retrain=False,
pre_file=pre_file,
old_epochs=old_epochs,
cross_index=cross_index)
spatial_a.summary()
spatial_b.summary()
spatial.summary()
# print(glob.glob('weights/inception_spatial2fc_{}_{}e_cr{}.h5'.format(n_neurons,pre_train[0],cross_index)[-1]))
spatial.load_weights('weights/sinception_spatial2fc_256-62-0.8554.hdf5')
print("Load")
spatial_a.layers[0].set_weights(spatial.layers[0].get_weights())
n_neurons = args.neural
dropout = args.dropout
pre_file = 'inception_spatial2fc_{}'.format(n_neurons)
if train & (not retrain):
weights = 'imagenet'
else:
weights = None
if args.fine == 1:
fine = True
else:
fine = False
result_model = models.InceptionSpatial(
n_neurons=n_neurons, seq_len=seq_len, classes=classes,
weights=weights, dropout=dropout, fine=fine, retrain=retrain,
pre_file=pre_file,old_epochs=old_epochs,cross_index=cross_index)
if (args.summary == 1):
result_model.summary()
sys.exit()
lr = args.lr
decay = args.decay
losses = {
"loss1": "categorical_crossentropy",
"loss2": "categorical_crossentropy",
"loss3": "categorical_crossentropy"
}
lossWeights = {"loss1": 1.0, "loss2": 1.0, "loss3": 1.0}